ES2799887T3 - Embryo evaluation - Google Patents

Abstract

A computer-implemented method for sorting IVF embryos to predict their developmental potential after transfer; the method comprising: obtaining values for a plurality of characteristics related to the morphological development of the embryos during an observation period; determine for the respective embryos a measure of whether or not the embryo has undergone a direct cleavage event and classify embryos determined to have undergone a direct cleavage event with a rank indicating a lower developmental potential than for embryos that they have not been determined to have experienced a direct splitting event; and for embryos not determined to have undergone a direct cleavage event, determining whether or not a measure of the duration of a predefined developmental stage for the embryo exceeds a predefined threshold duration and classifying the embryos for which it is determined that the duration of the predefined developmental stage exceeds the predefined threshold duration with a classification indicating a lower developmental potential than for embryos for which the duration of the predefined developmental stage is not determined to exceed the predefined threshold duration; and for embryos for which the predefined developmental stage duration is not determined to exceed the predefined threshold duration, determining whether a measure of a relative duration of a predefined first developmental stage for the embryo relative to a second predefined developmental stage The predefined developmental potential for the embryo is or is not outside a predefined range and classifying embryos for which the relative duration is outside the predefined range with a classification indicating a lower developmental potential than for embryos for which the relative duration is not. outside the predefined range.

Description

DESCRIPTION

Embryo evaluation

Background of the invention

Certain embodiments of the present invention relate to methods and apparatus for evaluating the developmental potential of embryos and in particular to classifying / scoring embryos according to their development potential.

Infertility affects more than 80 million people around the world. It is estimated that 10% of all couples experience primary or secondary infertility. In vitro fertilization (IVF) is a voluntary medical treatment that can be provided to a couple who might not otherwise conceive of the possibility of establishing a pregnancy and becoming parents. It is a process in which eggs (oocytes) are removed from a woman's ovaries and then fertilized with sperm in the laboratory. The embryos created in this process are then placed in the uterus for possible implantation. Between fertilization and transfer, embryos are normally stored in an incubation chamber of an incubator for 2-6 days, during which time they can be regularly monitored, eg by imaging, to assess their development. Conditions within the incubator, such as temperature and atmospheric composition, are controlled, generally in order to emulate conditions in the oviduct and uterus.

In a typical IVF cycle, multiple eggs from a single patient will be fertilized and the resulting embryos will be incubated. However, it is common that not all incubated embryos are transferred to the patient's uterus. This is to reduce the risk of multiple potentially dangerous births. Embryos will normally be selected for transfer based on an assessment of the developmental potential of the embryos that have been incubated. The embryos that are determined to have the greatest potential to progress to live birth will preferably be selected over other embryos in your cohort. Therefore, an important aspect of IVF treatment is to assess the developmental potential of the embryos that comprise a cohort, that is, to determine the quality of the embryo where the quality of the embryo is a prediction representing the probability that an embryo will implant. successfully develop in the womb after transfer and lead to the birth of a healthy baby.

A powerful embryo quality assessment tool that has been developed in recent years is fast-camera embryo imaging. Fast-camera embryo imaging involves imaging embryos during development. This can allow to establish the times of various developmental events, such as cell divisions and / or the presence or absence of other characteristics related to the development of an embryo, for example, with respect to cellular uniformity (homogeneity) in different stages, the appearance of pronuclei (PN) and the presence of multinucleation (MN).

These times and characteristics can sometimes be called morphokinetic / morphological parameters for the embryo. In this regard, the terms "morphokinetic" and "morphological" will generally be used herein interchangeably, although in some respects morphokinetic characteristics can be considered strictly a subset of morphological characteristics, specifically, those morphological characteristics specifically related to the times. Studies have shown how the times and durations of various embryonic developmental events and the presence or absence of various other developmental characteristics can be correlated with the developmental potential of an embryo.

Models for embryo selection (i.e. models to assess the developmental potential of an embryo) that take into account morphokinetic parameters can be constructed, evaluated and validated using known implantation data (DIC), thus embryos with DIC Positive embryos are embryos that are known to have subsequently implanted and DIC negative embryos are embryos that are known to have not subsequently implanted.

As an example of a simple model for evaluating the developmental potential of an embryo, it has been found that a relatively early time of division from one cell to two cells is an indicator of a good quality embryo. Other morphokinetic parameters, for example, the degree of synchrony in the two divisions when dividing from two cells to four cells, have also been found to be sensitive to the quality of the embryo. More generally, various approaches have been proposed to assess the developmental potential of an embryo from parameters related to the in vitro development of the embryo. Accordingly, one goal of fast-motion imaging is to establish values for various parameters related to the timing of various embryonic development events and / or other characteristics related to embryo development, for example, with respect to cellular uniformity (homogeneity). at different stages, the appearance of pronuclei (PN) and the presence of multinucleation (MN). Establishing values and characteristics related to embryonic development from a series of fast-camera images is sometimes called annotation.

Although various times and characteristics associated with embryonic development have been found to help provide quality indicators for the development of an embryo, the specific values for these that indicate a good quality embryo may be different for different embryos depending on the conditions in which That incubates the embryo and the way the various events are distributed. For example, one clinic could incubate embryos with an atmosphere of a certain percentage of oxygen and temperature, while another clinic could incubate embryos with an atmosphere of different percentage of oxygen and temperature. This may mean that the optimal time for a given morphological event in the development of an embryo may be different for different incubation conditions / clinics.

Figure 1 is a graph that represents this principle (this graph is very schematic and not based on actual data). Thus, Figure 1 shows an example of how the implantation probability, Imp%, could vary as a function of the timing of an arbitrary developmental event X (e.g., the duration of a particular cell cycle or the time of a particular division). The solid curve represents the variation in implantation probability as a function of time observed for X for embryos that have developed according to a first set of conditions, while the dashed curve represents the variation for embryos that have developed according to a second set of terms. For example, the solid curve can represent embryos incubated in a relatively low oxygen atmosphere, while the dashed curve can represent embryos incubated in a relatively high oxygen atmosphere. As another example, the solid curve could represent embryos that have been fertilized by intracytoplasmic sperm injection (ICSI), while the dashed curve could represent embryos that have been fertilized by in vitro fertilization (IVF). As yet another example, the two curves could represent embryos that have developed in different clinics. The two curves in Figure 1 systematically compensate for each other because embryos incubated under different conditions will generally develop at different rates in at least some respects.

Therefore, although Figure 1 shows that the time associated with the developmental event X can be used to identify embryos that have a relatively high probability of implantation for embryos that have developed under both sets of conditions, the actual values of X associated with high implantation probability are different for the two groups. For example, for embryos incubated in the first set of conditions (solid line), an optimal interval for the time of X could be considered to be from h1 to h2, while an optimal interval for embryos incubated in the second set of conditions could be considered. conditions (dashed line) is from h3 to h4. What this means in practice is that different models will be needed to assess the developmental potential of embryos for different populations.

However, it would be preferable to establish a single model that is applicable to embryos that have developed under a number of different conditions, that is, what might be called a universally applicable model (or at least one model applicable to embryos that have developed in a series different conditions). A universally applicable model would not only simplify the process of selecting a model for use in different conditions of embryonic development, in some cases there may not be enough DIC data available for a given set of development conditions to allow it to be reliably established. a model for those specific conditions, for example, in the case of a "new" clinic. A simple solution to provide a single model for the schematic situation depicted in Figure 1 would be to assume an optimal interval for X time between h1 and h4 for all embryos. However, this would result in the embryos in each population being wrongly classified as having a high probability of implantation, which, of course, is not a satisfactory solution.

Consequently, there is a desire to develop models to assess the developmental potential (viability / quality) of embryos, such as an in vitro incubation of human embryos, which are applicable for embryos that have developed under a number of different conditions.

Document US2014220618 discloses methods, compositions and equipment to determine the development potential of one or more embryos. The methods, compositions, and kits are said to find use in identifying embryos in vitro that are most useful in treating infertility in humans.

WO2012163363 discloses a method and system for selecting embryos for in vitro fertilization based on the timing and duration of observed cell divisions and associated cell morphology.

Document W02013004239 discloses a system and a method to determine quality criteria in order to select the most viable embryos after in vitro fertilization . The approach can further be applied to iteratively adapt embryo quality criteria based on new knowledge, historical selection and fertilization data, and cooperation between fertility clinics.

Document W02014033210 discloses a computer-implemented method to automatically detect variations and / or abnormalities in the development conditions of embryos that are incubated in vitro.

Meseguer et al in Human Reproduction, published for the European Society for Human Reproduction and Embryology by IRL Press, ISSN 0268-1161, 10-01-2011 disclose approaches to using morphokinetics as a predictor of embryo implantation.

Rubio et al in Fertility And Sterility, Elsevier Science Inc, New York, NY, United States, ISSN 0015-0282, 01-122012 report a fast-camera study related to the limited success of direct division human embryo implantation.

Ciray et al in Human Reproduction, Oxford Journals, UK, ISSN 0268-1161, 12-01-2014 propose guidelines on the nomenclature and annotation of dynamic monitoring of human embryos by a group of fast camera users.

Kaser et al in Human Reproduction Update Nov-Dec 2011, ISSN 1460-2369, 09-01-2014 discloses a review of clinical outcomes after preimplantation human embryo selection with fast-camera supervision.

Summary of the invention

According to a first aspect of the invention, there is provided a method for classifying IVF embryos to predict their developmental potential after transfer; the method comprising: obtaining values for a plurality of characteristics related to the morphological development of the embryos during an observation period; determine for the respective embryos a measure of whether or not the embryo has undergone a direct division event and classify embryos that have been determined to have undergone a direct division event with a classification indicating a lower developmental potential than for embryos that they have not been determined to have experienced a direct division event; and for embryos that have not been determined to have undergone a direct division event, determine whether or not a measure of the duration of a predefined developmental stage for the embryo exceeds a predefined threshold duration and classify the embryos for which it is determined that the duration of the predefined stage of development exceeds the predefined threshold duration with a rating indicating a lower developmental potential than for embryos for which the duration of the predefined stage of development is not determined to exceed the predefined threshold duration; and for embryos for which the duration of the predefined stage of development is not determined to exceed the predefined threshold duration, determining whether a measure of a relative duration of a predefined first stage of development for the embryo with respect to a second stage of development predefined development for the embryo is or is not outside a predefined range and classify embryos for which the relative duration is outside the predefined range with a classification indicating a lower developmental potential than for embryos for which the relative duration is not outside the predefined range.

According to a second aspect of the invention, there is provided an apparatus for classifying IVF embryos to predict their developmental potential after transfer, the apparatus comprising: a data input element configured to obtain values for a plurality of characteristics related to the morphological development of the embryos during a period of observation; and a processing element configured to: determine for the respective embryos a measure of whether or not the embryo has undergone a direct division event and to classify embryos that have been determined to have undergone a direct division event with a classification indicating a potential less developmentally than for embryos that have not been determined to have undergone a direct division event; and for embryos that have not been determined to have undergone a direct division event, determine whether or not a measure of the duration of a predefined developmental stage for the embryo exceeds a predefined threshold duration and classify the embryos for which it is determined that the duration of the predefined stage of development exceeds the predefined threshold duration with a rating indicating a lower developmental potential than for embryos for which the duration of the predefined stage of development is not determined to exceed the predefined threshold duration; and for embryos for which the duration of the predefined stage of development is not determined to exceed the predefined threshold duration, determining whether a measure of a relative duration of a predefined first stage of development for the embryo with respect to a second stage of development predefined development for the embryo is or is not outside a predefined range and classify embryos for which the relative duration is outside the predefined range with a classification indicating a lower developmental potential than for embryos for which the relative duration is not outside the predefined range.

Other aspects of the invention include a non-transient computer program product containing machine-readable instructions to carry out the method of the first aspect of the invention and a loaded and operational apparatus for executing machine-readable instructions to carry out the method of the first aspect of the invention.

Additional aspects and features of the invention are defined in the claims.

It will be appreciated that the features and aspects of the invention described herein in relation to the first and other aspects of the invention are equally applicable to, and may be combined with, embodiments of the invention according to other aspects of the invention as appropriate. and not just in the specific combinations described above.

It will be appreciated that the features and aspects of the invention described above in connection with the first and other aspects of the invention are equally applicable to, and may be combined with, embodiments of the invention according to other aspects of the invention as appropriate and not only in the specific combinations described previously.

Brief description of the drawings

The invention is now described by way of example only with reference to the accompanying drawings, in which:

Figure 1 is a highly schematic diagram depicting how implantation probability might vary as a function of time associated with an arbitrary developmental event for two populations of embryos incubated under two different sets of conditions;

Figure 2 schematically represents some nomenclature as used herein for an embryo division pattern showing division times (t2 to t5), cell cycle lengths (cc1 to cc3), and synchronies (s2 and s3) in relation to the images obtained;

Figure 3 schematically represents an embryo at different embryonic development events since initial insemination (at time t = 0) and at division times t2-t8 with some associated aspects of chronology terminology as used herein. ;

Figure 4 schematically represents an apparatus for determining a developmental potential for an embryo according to an embodiment of the invention;

Figure 5 is a flow chart schematically depicting a method for classifying embryos based on their developmental potential in accordance with some embodiments of the invention;

Figure 6 is a classification tree diagram schematically representing the application of the model of Figure 5 to 3,275 embryos having known implantation data (DIC embryos);

Figure 7 is a bar graph depicting the distribution of several of the 3,275 DIC embryos associated with the classification tree of Figure 6 among different scores established according to the method of claim 5;

Figures 8, 10 and 12 are classification trees schematically representing the application of other models to 3,275 DIC embryos according to other embodiments of the invention;

Figures 9, 11 and 13 are bar graphs representing the distribution of several of the 3,275 DIC embryos among the different scores established according to the models represented by the classification trees in Figures 8, 10 and 12 respectively;

Figures 14 to 17 are classification trees schematically representing the application in the same way to different subpopulations of the 3,275 DIC embryos according to embodiments of the invention; and Figure 18 represents for an analysis of 10,316 embryos for which it is known whether or not the embryos developed to the blastocyst stage 120 hours after fertilization, the proportion of embryos that developed to the blastocyst stage 120 hours later of fertilization for each of the different scores established for each of the embryos according to the model represented in figure 6.

Detailed description

Aspects and characteristics of certain examples and embodiments of the present invention are discussed / described herein. Some aspects and characteristics of certain examples and embodiments may be implemented in a conventional manner and these are not discussed / described in detail for brevity. Thus, it will be appreciated that aspects and features of the apparatus and methods discussed herein that are not described in detail may be implemented according to any conventional technique to implement such features and features.

Unless the context requires otherwise, terms used herein should be construed according to their meanings, as commonly understood by one of ordinary skill in the art to which this disclosure pertains. Some terms may be used herein in accordance with the following definitions (unless the context dictates otherwise).

The division time (cell division time) is defined as the first observed time point relative to a defined start point (zero time) when the newly formed blastomeres are completely separated by confluent cell membranes, the division time is, for hence the completion time of a blastomere division. The division times can therefore be defined as follows:

t2: division time to embryo of 2 blastomeres

t3: time of division to embryo of 3 blastomeres

t4: division time to embryo of 4 blastomeres

t5: division time to embryo of 5 blastomeres

t6: division time to embryo of 6 blastomeres

t7: division time to embryo of 7 blastomeres

t8: division time to embryo of 8 blastomeres

tn: division time to embryo of n blastomeres

It will be appreciated that the division time can be defined equally with respect to other associated time points. with cell division. For example, in the above definition, the division time is related to a time associated with the completion of cell division (that is, when the newly formed blastomeres are completely separated), but in another implementation, the division time can be defined equally as a time associated with the beginning of cell division or a time associated with an intermediate point for cell division.

In the present context, division times are usually expressed as hours after a defined zero time. Time zero may be the time of insemination (e.g., the time of intracytoplasmic sperm injection (ICSI), also called microinjection) or it could also be later than the time of mixing of sperm and oocytes (in IVF traditional) or after the moment when the successful fusion of gametes to form a new organism (the zygote) is observed for the first time, that is, exclusion of the second polar body. Similarly, it could be after the time of pronuclear onset or fading / disappearance or other significant developmental parameter. With respect to pronuclear fading / disappearance, the terms "faded" and "disappeared" in relation to pronuclei (PN) can be used interchangeably herein. The term tPNf can be used to indicate a certain time for the pronuclei to fade (that is, a certain time point such as the time when the pronuclei (PN) are no longer distinguished).

The duration of the first cell cycle cc1 is the period between fertilization and the division time t2 that provides the first pair of descendant cells (ie, the first second generation cells). The duration of the second cell cycle cc2 is the period between the division time t2 that provides the first pair of daughter cells and the division time t3 that provides the first pair of third descendant cells (i.e., the first third generation cells) . The duration of the third cell cycle cc3 is the period between the division time t3 that provides the first pair of third descendant cells and the division time t5 that provides the first pair of fourth descendant cells (that is, the first fourth generation cells ). The duration of the fourth cell cycle cc4 is the period between the division time t5 that provides the first pair of fourth descendant cells and the division time t9 that provides the first pair of fifth descendant cells (that is, the first fifth generation cells ).

These cell cycle durations are therefore based on the blastomere dividing faster for each new generation. However, there are additional cell cycle durations associated with slower blastomere division.

For example, in addition to the cc2 cell cycle duration, there is a cc2b cell cycle duration corresponding to the period between the division time t2 provided by the first pair of descendant cells and the division time t4 provided by the second pair of third cells. decendents. In this regard, the cc2 cell cycle duration may also be referred to as the cc2a cell cycle duration for simplicity of terminology.

Also, in addition to the duration of the cc3 cell cycle, there is a duration of the cc3b cell cycle corresponding to the period between the division time t3 provided by the first pair of descendant cells and the division time t6 provided by the second pair of fourth descendant cells. . There is also a cell cycle duration cc3c corresponding to the period between the division time t4 provided by the second pair of third descendant cells and the division time t7 provided by the third pair of fourth descendant cells. There is also a cell cycle duration cc3d corresponding to the period between the division time t4 provided by the second pair of third descendant cells and the division time t8 provided by the fourth pair of fourth descendant cells. In this regard, the cc3 cell cycle duration may also be referred to as the cc3a cell cycle duration for consistency in terminology.

Therefore, the duration of cell cycles is defined as follows:

• cc1 = t2: First cell cycle.

• cc2 (also called cc2a) = t3-t2: Second cell cycle, duration of the period as a 2-blastomere embryo.

• cc2b = t4-t2: Second cell cycle for both blastomeres, duration of the period as an embryo of 2 and 3 blastomeres.

• cc3 (also called cc3a) = t5-t3: third cell cycle, duration of the period as an embryo of 3 and 4 blastomeres.

• cc3b, cc3c, cc3d = t6-t3; t7-t4; and t8-t4 respectively: Third cell cycle for slower blastomeres, duration of the period as an embryo of 3, 4 and 5 blastomeres; as a 4, 5 and 6 blastomere embryo and as a 4, 5, 6 and 7 blastomere embryo respectively.

• cc2_3 = t5-t2: Second and third cell cycle, duration of the period as an embryo of 2, 3 and 4 blastomeres (ie, cc2 cc3).

• cc4 = t9-t5: Fourth cell cycle, duration of the period as an embryo of 5, 6, 7 and 8 blastomeres.

Synchronicities are defined as follows:

• s2 = t4-t3: Synchrony in division from embryo of 2 blastomeres to embryo of 4 blastomeres.

• s3 = t8-t5: Synchrony in division from embryo of 4 blastomeres to embryo of 8 blastomeres.

• s3a = t6-t5; s3b = t7-t6; s3c = t8-t7: Duration of individual cell divisions involved in development from a 4 blastomere embryo to an 8 blastomere embryo.

Figures 2 and 3 schematically represent some aspects of the terminology used herein with respect to the times and durations of some embryonic development events, as discussed above. Figure 2 shows various images of an embryo at various stages of development and indicates various times associated with various developmental events, such as t2, t3, t4, t5, cc1, cc2 (which may also be referred to herein as cc2a). , cc3 (which may also be referred to herein as cc3a), s2, and s3. Figure 3 schematically represents from left to right the development of the embryo through the stages of one, two, three, four, five, six, seven and eight blastomeres. The times t2 to t8 in which the respective cell division stage is completed are marked schematically along the lower axis. Figure 3 also indicates schematically the durations of the cc1, cc2a, cc2b, cc3a, cc3b, cc3c and cc3d cell cycles and S2 and S3 synchronicities.

The division period is defined as the period of time from the first observation of clefts in the cell membrane (indicating the start of cytoplasmic division) until the division of the cytoplasmic cells is complete so that the blastomeres are completely separated. by confluent cell membranes. Also called duration of cytokinesis.

Fertilization and division can be considered in some respects the major morphological / morphokinetic events of an embryo, at least up to the 8 blastomere stage or until the onset of compaction. The division time, cell cycle, division synchrony, and the division period are examples of embryonic morphological parameters that can be defined from these primary morphological events and each of these embryonic morphological parameters can be defined as the duration of a period. time between two morphological events, e.g. eg, measured in hours.

Direct division is when a cell divides faster than the time it is supposed to take for its DNA to replicate properly before division. For example, a cell may divide into two descendant cells, and then if one of the descendant cells divides into two descendant third cells before the time it normally takes for its DNA to replicate, it can be assumed that the initial cell has experienced an event. direct division. That is, the initial cell has divided, within a predefined period of time, from one cell to more than two cells, e.g. eg, three cells. Direct division can occur in different cell cycles. For example, referring to Figure 3, a short period for any of cc2a, cc2b, cc3a, cc3b, cc3c, cc3d can be interpreted as indicative of a direct split event.

As already mentioned, it is known that it establishes a measure of the development potential of an embryo from various parameters associated with its development, such as parameters corresponding to (or based on) the times analyzed above and, to do this, it is can determine the values for the relevant parameters of interest from fast-camera images of the embryo as it develops through the relevant stages. In some approaches to determining the developmental potential of an embryo, other developmental characteristics may be of interest. For example, an evaluation of the quality of an embryo can take into account the values established for the following morphological characteristics:

• NOT2PN: Indication of whether or not two pronuclei are adequately identified for the embryo. This characteristic can be determined visually from an image of the embryo at the appropriate developmental site with a mental stage and can, for example, take a single binary value to indicate whether or not pronuclei are properly identified in the embryo, or it could take a value indicating the number of targets identified for the embryo, for example, a value corresponding to "0", "1", "2", "3" or "4 or more" depending on the number of pronuclei identified for the embryo ( a value of "2" is normal).

• MN2: Indication of (any) multinucleation observed at the two-blastomere (cell) stage. This characteristic can be determined visually from an image of the embryo at the appropriate stage of development and can take values corresponding to "0", "1" or "2" corresponding to the number of cells that have been determined to show multinucleation at the stage. of two blastomeres.

• MN4: Indication of (any) multinucleation observed at the four blastomere stage. This characteristic can be determined visually from an image of the embryo at the appropriate stage of development and can take values corresponding to "0", "1", "2", "3" or "4" corresponding to the number of cells that are has identified that they show multinucleation at the four blastomere stage.

• UNEVEN2: Indication of (ir) regularity of blastomeres in the two-blastomere stage. This characteristic can be determined visually from an image of the embryo at the appropriate stage of development and can take values corresponding to "regular" (the blastomeres in the embryo of two blastomeres are classified as regular) or "irregular" (the blastomeres in the embryo of two blastomeres are classified as irregular).

• UNEVEN4: Indication of (ir) regularity of blastomeres in the four blastomere stage. This characteristic can be visually determined from an image of the embryo at the appropriate stage of development and can take values corresponding to "regular" (blastomeres in the four-blastomere embryo are classified as regular) or "irregular" (blastomeres in the four-blastomere embryo are classified as irregular).

It will be appreciated that setting values for some of these parameters may include a component of subjectivity, for example, as to whether the cells comprising an embryo are regular or not. It will also be appreciated that the terminology adopted for the specific values (eg "regular", "irregular") is not meaningful and the values could also be characterized in other ways as well, eg. eg, as "true" or "false" by numerical values associated with the different potential states, eg. eg, "0" for regular, "1" for irregular).

It will be further appreciated that the above identified times and characteristics associated with embryonic development are established to provide a general understanding of the types of parameters and characteristics that may be of interest when attempting to provide models for assessing the developmental potential of embryos and typically only a subset of these parameters and / or characteristics will be of interest for a given model.

Embryo quality is a measure of an embryo's ability to implant and develop successfully in the uterus after transfer. High-quality embryos have a greater chance of implanting successfully (high probability of implantation) and developing in the uterus to a healthy baby after transfer than low-quality embryos. However, even a high-quality embryo is not a guarantee of implantation, as the actual transfer and receptivity of the woman influences the final result.

Feasibility and quality can be used interchangeably. The measurement of embryo quality (or viability) is a parameter intended to reflect the quality (or viability) of an embryo in such a way that embryos with certain values of the quality parameter (eg, high or low values depending on of how the parameter is defined) have a high probability of being of high quality (or viability) and a low probability of being of low quality (or viability). On the other hand, embryos with other values determined for the quality parameter (or viability) have a low probability of having high quality (or viability) and a high probability of being of low quality (or viability).

The term "developmental potential" can be used to reflect an estimated probability that an embryo will develop to the blastocyst stage, implant, lead to pregnancy, develop to a stage associated with a heartbeat, and / or lead to a baby born alive. In some embodiments, the developmental potential may be a determination of the quality of the embryo. The potential for development can be equated with the quality of the embryo. An embryo that has a positive development potential (i.e. good (high) embryo quality) is one that is more likely to develop to the blastocyst stage and / or lead to successful implantation and / or develop in the embryo in the uterus after transfer and / or gives rise to pregnancy and / or gives rise to a live born baby compared to an embryo that has negative development potential (or poor (low) embryo quality).

Thus, it is determined that embryos that have been determined to be of good (high) quality have a greater chance of successfully implanting and / or developing in the uterus after transfer compared to poor quality embryos. However, it will be appreciated that a high quality embryo is not a guarantee of implantation, as the actual transfer and receptivity of the woman greatly influences the end result.

In some cases, the term "embryo" can be used to describe a fertilized oocyte after implantation in the uterus for up to 8 weeks after fertilization, when it develops into a fetus. By this definition, the fertilized oocyte is often referred to as a pre-embryo or zygote until implantation occurs. However, the term "embryo" as used herein will have a broader definition, including the pre-embryonic phase. The term "embryo" as used herein encompasses all stages of development from fertilization of the oocyte to morula, the blastocyst, hatching, and implantation stages. Accordingly, the term embryo may herein denote each of the stages of the fertilized oocyte, zygote, 2-cell, 4-cell, 8-cell, 16-cell, compaction, morula, blastocyst, expanded blastocyst, and hatched blastocyst, as well as all intermediate stages (eg, 3 cells or 5 cells).

An embryo is roughly spherical and is composed of one or more cells (blastomeres) surrounded by a gelatin-like covering, the acellular matrix known as the zona pellucida. The zona pellucida performs various functions until the embryo hatches and is a good point of reference for evaluation of the embryo. The zona pellucida is spherical and translucent and should be clearly distinguished from cell debris.

An embryo is formed when an oocyte is fertilized by fusion or injection of a sperm cell (sperm). The term embryo is traditionally used also after hatching (that is, rupture of the zona pellucida) and subsequent implantation. For humans, the fertilized oocyte is traditionally called a zygote or embryo during the first 8 weeks. After that (that is, after eight weeks and when all the major organs have formed) it is called a fetus. However, the distinction between zygote, embryo, and fetus is not generally well defined. The terms embryo and zygote can be used interchangeably herein.

An embryo that is evaluated in accordance with embodiments of the invention as described herein it can be previously frozen, e.g. eg, embryos cryopreserved immediately after fertilization (eg, at the 1-cell stage) and then thawed. Alternatively, they can be freshly prepared, e.g. g., embryos freshly prepared from oocytes using IVF or IICE techniques, for example. It will be appreciated that, to the extent that development of an embryo has been frozen, the times of developmental events after fertilization can be defined by ignoring the time between freezing and thawing. Alternatively, an onset time can be defined as one of the first developmental events, such as the exclusion of the second polar body or the appearance / disappearance of pronuclei, after thawing.

Fertilization can be considered to be the moment when the oocyte recognizes and accepts the sperm. The sperm triggers the activation of the ovum after the meiotic cycle of the oocyte has been suspended in the metaphase of the second meiotic division. This results in the production and extrusion of the second polar body. A few hours after the fusion of the sperm and the egg, DNA synthesis begins. Male and female pronuclei (PN) appear. The NPs move to the center of the egg and the membranes break down and the NPs disappear (vanish). This combination of the two genomes is called syngamy. Next, cell divisions begin.

The moment when the pronuclei disappear may be referred to as tPNf. The terms "fade" or "faded" and "disappear" or "disappeared" in relation to the pronuclei (PN) may be used interchangeably herein.

During embryonic development, the numbers of blastomeres increase in geometric progression (1-2-4-8-16-etc.). Synchronous cell division is generally maintained at the 8-cell stage or later, until compaction in human embryos. After that, cell division becomes asynchronous and eventually individual cells have their own cell cycle. Human embryos produced during infertility treatment can be transferred to the recipient before the 8 blastomere stage. In some cases, human embryos are also cultured to the blastocyst stage prior to transfer. This is preferably done when there are many good quality embryos available or long incubation is necessary to await the result of a preimplantation genetic diagnosis (PGD). However, there is a trend towards long incubation as incubation technology improves.

Some illustrative implementations of embodiments of the invention can be used to establish blastocyst-related parameters.

A blastocyst quality / measure criterion is an example of an embryo quality / measure criterion. The quality criteria of the blastocysts can, for example, be related to the development of the embryo from compaction, that is, initial compaction, to the hatched blastocyst. Compaction is a process in which an intensification of the contacts between tightly bonded blastomeres and desmosomes leads to a reduction of the intercellular space and a blurring of the cell contours. Before compaction, the embryo's blastomeres can be tracked individually and before compaction, embryonic development follows a route of defined and mostly synchronous cell divisions that can be easily observed and annotated. After compaction, embryonic development is characterized by a more or less continuous development from morula to blastocyst, where individual blastomeres become difficult to trace, but various stages can nonetheless be characterized by setting values for parameters associated with these stages by means of visual inspection of the images obtained for the relevant development stages.

The onset of compaction (IC) describes the first time a compaction is observed between two or more blastomeres. Therefore, the IC marks the beginning of the compaction process.

The morula (M) is associated with the first time that no plasma membranes are visible between the blastomeres. When the compaction process is complete, no plasma membranes are visible between any of the blastomeres that form the compaction, and the embryo can be defined as a morula. Most often, the morula is seen after the third S3 sync period (i.e. after t8) near, or right at the beginning, of the fourth S4 sync period (i.e. at t9), but may be earlier. Infrequently, embryos divide into 16 cells or more before compaction begins in human embryos.

Initial trophectoderm differentiation (IDT) is defined as the first time defined trophectoderm cells are recognized. Blastulation Onset (IB) is defined as the first time a fluid-filled cavity, the blastocele, can be seen. Also called "cavitation onset." It describes the beginning of the transition period between the morula stage and the blastocyst stage of the embryo. Embryos often remain in this transitional stage for a period of time before entering the actual blastocyst stage. The onset of cavitation usually appears immediately after the differentiation of the trophectoderm cells. The outer layer of the morula in contact with the outside environment begins to actively pump salt and water into the intercellular space, as a result of which a cavity (the blastocele) begins to form.

The initial differentiation of the inner cell mass (DIMCI) defined as the first time that the inner cell mass can be recognized. DIMCI describes the initiation of inner cell mass development. A group of cells placed in eccentric position connected by communicating junctions where the boundaries between cells do not appear to be well defined.

The blastocyst (B) can be defined as the last image before the blastocyst begins to expand. When this happens, the zona pellucida usually begins to change and there is a clear distinction between trophectoderm and cells of inner cell mass.

The onset of blastocyst expansion (BE) can be defined as the first time that the embryo has filled the perivithelial space and begins to move / expand the zona pellucida. EB can describe the beginning of the expansion of the embryo. As the blastocyst expands, the zona pellucida becomes visibly thinner.

Blastocyst hatching (EcB) can be defined as the first time that a trophectoderm cell has escaped / penetrated the zona pellucida or a certain fraction has hatched.

Fully hatched blastocyst (CE) is defined as when hatching is completed with detachment of the zona pellucida.

Various times associated with blastocyst development can be defined as follows:

tM = Time from insemination to morula formation (hours)

tSB = Time from insemination to start of blastulation (hours)

tB = Time from insemination to blastocyst formation (hours)

tEB = Time from insemination to expanded blastocyst formation (hours)

tHB = Time from insemination to blastocyst hatching (hours)

Figure 4 schematically represents an apparatus 2 for determining a developmental potential for an embryo 8 according to certain embodiments of the invention. Apparatus 2 comprises a general purpose computer 4 coupled to an embryo imaging system 6. The embryo imaging system 6 may be generally conventional and is configured to image the embryo 8 at various stages of development in accordance with established techniques. It will be appreciated that, in general, the embryo imaging system 6 will typically be configured to image a plurality of embryos, rather than a single embryo, during a monitoring period. For example, a typical study may involve the analysis of several embryos, for example, 72 embryos. The embryo imaging system can be configured to record images of each embryo (potentially with images that are taken in multiple focal planes) one at a time before imaging the next embryo. Once all the embryos have been imaged, which could take, for example, 5 minutes, the cycle of imaging the individual embryos can be repeated to provide respective images for the respective embryos for the next time point.

The general purpose computer 4 is adapted (programmed) to execute a method for determining / evaluating a developmental potential of an embryo from an analysis of images obtained from the embryo imaging system 6 as further described below.

Thus, the computer system 4 is configured to perform embryo image data processing in accordance with one embodiment of the invention. The computer 4 includes a central processing unit (CPU) 24, a read-only memory (ROM) 26, a random access memory (RAM) 28, a hard disk drive 30, a hardware interface 46, a controller display 32 and a display 34 and a user input / output (IO) circuit 36 with a keyboard 38 and a mouse 40. These devices are connected via a common bus 42. The computer 4 also includes a graphics card 44 connected via common bus 42. The graphics card includes a graphics processor (GPU) and random access memory tightly coupled to the GPU (GPU memory). The embryo imaging system 6 is communicatively coupled to the computer 4 through the hardware interface 46 in accordance with conventional techniques.

CPU 24 may execute program instructions stored within ROM 26, RAM 28, or hard drive 30 to perform embryo image data processing that can be stored within RAM 28 or hard drive. 30. RAM 28 and hard disk drive 30 are collectively called system memory. In some implementations, processing in accordance with embodiments of the invention may be based on images of embryos obtained by computer 4 directly from image capture system 6. In other implementations, the processing according to embodiments of the invention may be based on embryo images previously obtained and stored in a memory of the computer 4, e.g. eg, in RAM 28 of HDD 30 (ie, embryo imaging system 6 itself is not a necessary element of embodiments of the invention). Aspects of computer 4 can be largely conventional, except that the CPU is configured to run a program, which, for example, can be stored in RAM 28, r Om 26, or HDD 30, to perform processing according to certain embodiments of the invention as described herein.

The embryo 8 according to certain illustrative implementations is regularly monitored using the embryo imaging system 6 to obtain the relevant information (i.e., the times associated with particular embryonic developmental events, appearance (or not) of particular characteristics of the embryonic development). The embryo is preferably monitored at least once an hour, such as at least twice an hour, such as at least three times an hour, such as at least four times an hour, such as at least six times an hour, such as at least 12 times per hour. Monitoring is preferably done while the embryo is located in an incubator used to grow the embryo. This is preferably accomplished through image acquisition of the embryo, as discussed herein in connection with fast-camera methods.

Figure 5 is a flow chart schematically depicting a method for classifying embryos according to their developmental potential in accordance with certain embodiments of the invention. The method can be applied, for example, to classify a plurality of embryos associated with a single patient to help identify which of the embryos are most likely to implant successfully / lead to live birth. The number of embryos will, of course, vary between patients and treatment cycles, but in a typical case, the plurality of embryos from a single patient in a single treatment cycle could be between 6 and 12 embryos, for example. The method may be a computer-implemented method that can be implemented using the computer 4 of FIG. 4 with the CPU 24 implementing the method according to a loaded program.

In stage S1, a plurality of values are obtained for the characteristics related to the morphological development of the embryos during an observation period. In an approach to classifying embryos in the context of evaluating the embryos to determine their developmental potential relevant to a day three transfer, this observation period may, for example, extend to about 72 hours, or more, after relevant reference time (zero time). The morphological features of interest will depend on the specific application in question, as different embodiments may depend on different morphological features, as discussed further below. In this example, the characteristics obtained for each embryo are assumed to be:

(i) an indication of whether or not the embryo adequately presents two pronuclei (eg, a value for NOT2PN as defined above).

(ii) values for t3 and tPNf (considering that a relatively small difference between these values is an indicator of the appearance of direct division)

(iii) the duration of a predefined developmental stage that is used here to identify whether the development of the embryo is inappropriately slow (in this example, the developmental stage is from time zero (eg, IICE microinjection time) to t3)

(iv) the durations of two predefined stages of development used here to identify whether the relative duration of one of these stages of development to the other is within a desired range (in this example, the two stages of development are the duration of cc3a (ie, t5 -t3) and the combined duration of cc2a and cc3a (ie, t5-t2)). (v) an indication of whether the number of cells in the embryo failed to reach a given number within a given time (which, in this particular example, is eight cells in 66 hours).

Thus, to summarize, values for the following characteristics related to embryo morphological development can be sought for each embryo in accordance with an illustrative implementation of the method of Figure 5: NOT2PN; tPNf; t2; t3; t5 and t8. It will be appreciated that it may not be possible to obtain values for all of these characteristics for all embryos. For example, it may be that an embryo does not reach the 8 cell stage within the observation period, in which case it would not be possible to obtain a value for t8 (corresponding to a determination that the embryo did not reach 8 cells within the observation period. /supervision).

Values for these parameters can be obtained according to conventional techniques, for example, by using conventional fast-camera imaging of embryos in an incubator to obtain a series of images of developing embryos and using standard annotation procedures to identify the times and the appearances of the various relevant morphological events of the series of images. For example, the values for these parameters can be obtained using an EmbryoScope (RTM) device for fast-camera monitoring of embryos during incubation and its associated EmbryoViewer (RTM) software to annotate events of interest. The EmbryoScope (RTM) device and EmbryoViewer (RTM) software have been developed by, and are available from, Unisense FertiliTech A / S (Aarhus, Denmark). It will be appreciated that according to previously established techniques, annotation can be performed manually (eg, based on user input) and / or automatically (eg, based on numerical analysis / processing of the image) and / or semi-automatic (eg, based on a mixture of numerical image processing and user input).

At stage S2, one of the embryos is selected for consideration. In general, the approach in Figure 5 is to sequentially establish a score for each embryo and the score can then be used as the basis for ranking the embryos according to their developmental potential. The order in which the embryos are considered is not significant and, in this regard, the embryo that is selected at stage S2 can be arbitrarily chosen from the embryos that remain to be scored. The embryo selected for a given iteration can be called the current embryo for that iteration.

At step S3, a determination is made as to whether the current embryo should be classified. In this example, this is based on the value of NOT2PN. In some respects, this can be seen as an initial screening stage in which embryos are identified that have one or more characteristics that are known to be strongly associated with low developmental potential.

If it is determined at stage S3 that the current embryo does not adequately present two pronuclei, it is determined that the embryo should not be further classified and processing follows the branch marked N until stage S4 where the current embryo is assigned a score of 0 and processing continues to step S16. At step S16 it is determined whether the entire plurality of embryos has been considered or if there are more embryos to consider. If there are more embryos to consider, processing follows the S-marked branch back to stage S2 when the next embryo is selected for consideration.

If, on the other hand, it is determined at stage S3 that the embryo does present adequately two pronuclei, processing follows the branch marked S until stage S5.

At step S5 an evaluation is performed to determine if the embryo underwent direct division during the observation period (direct division has been found to be associated with relatively low developmental potential). In this example, this is based on determining whether a period of time associated with the development of the embryo is less than a threshold duration and, if so, determining that the embryo has undergone a direct division event. In this illustrative implementation, a direct split event is considered to have occurred if a measure of time between tPNF and t3 (t3-tPNf) is less than about 11.5 hours (eg, less than 11.48 hours ). Other threshold values could be used, for example, the threshold value can be selected from the group comprising: 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours, 11 hours, 12 hours, 13 hours and 14 hours.

In some other illustrative implementations, different time periods can be used to identify whether there has been a direct split event. For example, a direct split event can be considered to have occurred if a measure of the period (t3-t2) is less than a threshold quantity, for example, a quantity selected from the group comprising: 1 hour, 2 hours, 3 hours, 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours and 11 hours. In yet another example, a direct split event may be considered to have occurred if a measure of the period (t5-t4) is less than a threshold quantity, for example, a quantity selected from the group comprising: 0.1 hours, 1 hour, 2 hours, 3 hours, 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours and 11 hours. In yet another example, a direct division event can be considered to have occurred if the ratio of a 3-cell division time, t3, to a 4-cell division time, t4 is less than a quantity threshold selected from the group comprising: 0.9, 0.8, 0.7 and 0.6. In some implementations, step S5 may involve determining if any of these time periods is less than its corresponding threshold to identify if there have been any direct split events.

If it is determined at step S5 that the current embryo has undergone direct division, processing follows the S-marked branch to step S6 where the current embryo is assigned a score of 1 and processing then continues to step S16. Otherwise, the processing follows the branch marked N to step S7.

In step S7, an evaluation is made to determine whether the duration of a predefined development stage is greater than a predefined threshold duration. More generally, at stage S7, an evaluation is performed to determine whether an aspect of embryo development indicates relatively slow development (which has been found to be associated with relatively low developmental potential). In this illustrative implementation, the predefined developmental stage duration is the time it takes to reach 3 cells, t3, and the predefined threshold duration is approximately 43 hours (eg, 42.91 hours). However, other threshold values could be used with respect to t2, for example, the threshold value can be selected from the group comprising: 38 hours, 39 hours, 40 hours, 41 hours, 42 hours, 43 hours, 44 hours, 45 hours, 46 hours, 47 hours and 48 hours.

In some other illustrative implementations, different time periods can be used to identify if there has been slow development. For example, slow development can be considered to have occurred if a measure of t2 is less than a threshold quantity, for example, a quantity selected from the group comprising: 27 hours, 28 hours, 29 hours, 30 hours, 31 hours, 32 hours, 33 hours, 34 hours and 35 hours. In yet another example, slow development may be considered to have occurred if a measure of t4 is less than a threshold quantity, for example, a quantity selected from the group comprising: 41 hours, 42 hours, 43 hours, 44 hours, 45 hours , 46 hours, 47 hours, 47 hours, 48 hours, 49 hours and 50 hours. In yet another example, slow development may be considered to have occurred if a measure of t5 is less than a threshold quantity, for example, a quantity selected from the group comprising: 55 hours, 56 hours, 57 hours, 58 hours, 59 hours , 60 hours, 61 hours, 62 hours and 63 hours. In some implementations, step S7 may involve determining whether any of several of these times is greater than the corresponding threshold values to identify if there has been slow development with respect to any of the time periods.

If it is determined at stage S7 that the current embryo has undergone slow development, processing follows the S-marked branch until stage S8 where the current embryo is assigned a score of 2 and processing then continues to step S16. Otherwise, the processing follows the branch marked N to step S9.

Steps S9 and S11 in effect operate together to determine whether the relative duration of a predefined first developmental stage for the embryo relative to a second predefined developmental stage for the embryo is within or outside of a predefined range. More generally, these stages are combined to assess whether aspects of embryonic development indicate irregular development with respect to the relative durations of two developmental stages compared to associated embryos with good developmental potential. In this example, the implementation of the evaluation is done in two stages, namely, determining whether the relative duration from the first stage to the second stage is below a lower limit for the interval in stage S9 and determining whether the duration Relative from the first stage to the second stage is above an upper limit for the interval in stage s11.

Thus, in step S9 an evaluation is performed to determine whether the relative duration of a predefined first developmental stage for the embryo with respect to a second predefined developmental stage for the embryo is below a predefined limit. In this illustrative implementation, the first predefined developmental stage is from division to 3 cells, t3, to division to five cells, t5, (i.e., cc3a) and the second predefined developmental stage is the stage from division to 2 cells, t2, until division into five cells, t5, (ie, cc2a plus cc3a). Therefore, a measure of relative duration used in this example is (t5-t3) / (t5-t2) (equivalent to cc3a / (cc2a cc3a)). A lower limit for this relative duration in this specific implementation is approximately 0.34 hours. However, other lower limit values could be used for this parameter, eg 0.1; 0.2; 0.3; 0.4 or 0.5. In some other illustrative implementations, different relative durations can be used to identify if there has been uneven development and some examples of this are discussed below.

If it is determined at step S9 that the current embryo has undergone irregular development because the determined relative duration is below the lower limit, the processing follows the branch marked S until step S10 where the current embryo is assigned a score of 3 and processing then continues to step S16. Otherwise, the processing follows the branch marked N to step S11.

At step S11 an evaluation is performed to determine whether the relative duration of a predefined first developmental stage for the embryo relative to a second predefined developmental stage for the embryo is above a predefined limit. In this illustrative implementation, the first and second stages of development are the same as those used in step S9 and the upper limit for this relative duration in this specific implementation is about 0.58 hours. However, other upper limit values could be used for this parameter, eg 0.6, 0.7, 0.8 or 0.9. As noted above, in some other illustrative implementations, other relative durations can be used to identify if there has been uneven development and some examples of this are discussed below.

If it is determined at step S11 that the current embryo has undergone irregular development because the relative duration determined for the relevant developmental stages is above the upper limit, the processing follows the branch marked S until step S12 where it is assigned the current embryo a score of 4 and processing then continues to step S16. Otherwise, the processing follows the branch marked N to step S13.

Therefore, and as already discussed above, it will be appreciated that the combination of steps S9 and S11 in effect provides a determination of whether the relative duration of a predefined first developmental stage for the embryo with respect to a second developmental stage The predefined range for the embryo is within a predefined range (in which case the processing of Figure 5 for the relevant embryo will reach stage S13) or outside the range, in which case the embryo will be assigned a score of 3 or 4, depending whether it is below or above the range. In this particular illustrative implementation, the relative duration corresponds to (t5-t3) / (t5-t2) and the predefined range is approximately 0.34 to 0.58. However, other ranges may be used, for example, the range may be selected from the group comprising: 0.1 to 0.9, 0.2 to 0.8, 0.3 to 0.7, 0.4 to 0 , 6 and 0.5 to 0.6. In still other examples, different predefined first and second developmental stages can be used to establish whether an embryo exhibits irregular development in stages corresponding to stages S9 and S13. For example, other parameters and ranges that could be used include:

• (t3-t2) / (t5-t2) - with a range of: 0.1 to 0.9, 0.1 to 0.8, 0.2 to 0.7, 0.3 to 0.6 or 0.4 to 0.5

• (t3-t2) / (t5-t3) - with a range of: 0.05 to 10, 0.1 to 9, 0.15 to 8, 0.2 to 7, 0.25 to 6, 0, 3 to 7, 0.35 to 6, 0.4 to 5, 0.45 to 4, 0.5 to 3, 0.6 to 2, or 0.75 to 1

• (t5-t3) / t5) - with a range of: more than 0.1, more than 0.2 and more than 0.3

• (t4-t3) / (t3-t2)) - with a range of more than 0.1, less than 0.2, less than 0.3, less than 0.4 or less than 0.5

• (t8-t5) / (t5-t3) - with a range of: less than 0.1, less than 0.15 or less than 0.2

• ((t3-t2) + (t5-t4)) / (t8-t4) - with a range of: more than 0.3, more than 0.4, more than 0.5, more than 0.6, more than 0.7 or more than 0.8

• (t8-t5) / (t8-t4) - with a range of: more than 0.3, more than 0.4, more than 0.5, more than 0.6, more than 0.7, more than 0.8, more than 0.9 or more than 0.97

• (t3-tPNf) / (t4-tPNf) (or t3 / t4) - with a range of: more than 0.35, more than 0.45, more than 0.55, more than 0.65, more than 0.75, more than 0.85, more than 0.95

• (t4-t3) / (t4-t2) - with a range of: less than 0.3, less than 0.4, less than 0.5, less than 0.6 or less than 0.7 • (t8-t5) / (t8-t2) - with a range: less than 0.2, less than 0.3, less than 0.4, less than 0.5, and less than 0.6.

It will be noted that some of the ranges listed above are joined only at one extreme (that is, some ranges are specified as a value greater than X). In this case, the processing of Fig. 5 can be modified to avoid one or the other of step S9 or S11 depending on whether the interval is greater than a given limit or less than a given limit. Alternatively, an arbitrary extreme value could be set for the relevant limit, such as 0 hours or 999 hours.

It will be appreciated that for all the different proportions of durations for various developmental events (which individually may comprise the sum of 27 amendments, such as for the parameter ((t3-t2) + (t5-t4)) / (t8-t4) ), what is significant is an indication of the relevant proportion. Although this can be calculated as stated above, it could nonetheless be determined as an "inverse" of these proportions. In this regard, it will be appreciated that an indication of a value for A / B is also an indication of a value for B / A.

In step S13 an evaluation is made to determine whether or not the current embryo developed to a predefined number of cells within a predefined time. In this particular illustrative implementation, step S13 is based on determining whether or not the current embryo achieved 8 cells in a 66 hour period. However, different parameters can be used for this aspect of the processing of Figure 5 in accordance with other illustrative limitations. For example, instead of 66 hours, the evaluation can be based on a different time period, for example, a time period selected from the group comprising: 64 hours, 65 hours, 66 hours, 67 hours, 68 hours, 69 hours , 70 hours, 71 hours and 72 hours.

If it is determined at step S13 that the current embryo failed to reach the required number of cells within the relevant time, the processing follows the branch marked with N until step S14 where the current embryo is assigned a score of 4 and the Processing then continues to step S16. Otherwise, the processing follows the branch marked S to step S15.

At step S15, the current embryo is assigned a score of 5 and processing continues to step S16.

As noted above, at step S16 it is determined whether there are any more embryos to consider from the plurality of embryos to be classified and, if so, processing follows the branch marked S back to step S2, where the Processing described above is repeated for the next embryo to be selected. If, on the other hand, there are no more embryos to consider (ie, a score has been assigned to the entire plurality of embryos to be classified), the processing follows the branch marked with N until step S17.

In step S17, the plurality of embryos are classified among themselves according to their respective scores. Basically, a higher score is considered to be indicative of greater potential for development. One or more embryos from the plurality of embryos that have been classified for implantation / transfer to a patient can then be selected based on their classification. For example, higher ranked embryos may be selected for implantation / transfer to a patient in preference to lower ranked embryos (lower score). In this regard, it will be appreciated that the specific numerical scores assigned to each embryo (ie, 0, 1,2, ... 5) are not significant. Other values could be used as well. Also, the score could be based on a low score indicating good development potential, in which case stage S4 of the processing of Figure 5 would be associated with a higher score than stage S6, which in turn would be associated with a higher score. than step S8, and so on. In fact, the scores need not be numerical. For example, embryos may be associated with an A score at stage S4, a B score at stage S6, and so on up to an E score at stage D15, and the classification may be based on the identification of embryos associated with the former. letter of the alphabet as relatively low developmental potential.

Thus, the processing of Figure 5 represents a method for classifying a plurality of embryos according to their developmental potential in accordance with one embodiment of the invention. This involves establishing whether or not various stages of development meet various criteria as discussed above. In particular, embryos are evaluated for direct division (stage S5), slow development (stage S7), and irregularity (stages S9 and S13 combined). The inventors have discovered that these particular characteristics, and their relative importance in establishing a general classification according to the process described above, provide a model for classifying the embryos of a cohort to identify those that are most likely to implant successfully and subsequently , lead to a live birth. Furthermore, the inventors have discovered that the approach is relatively insensitive to characteristics associated with the conditions of development of the embryo, such as the nature of the fertilization methods used (ICSI or classical IVF) and the nature of the incubation atmosphere (p eg, low oxygen or ambient oxygen).

To demonstrate the ability of figure 5 processing to successfully classify embryos according to their developmental potential, the model can be applied to historical data of embryos for which there are known implantation data (i.e., DIC embryos) to establish how much Well the model predicts the known outcomes for DIC embryos. The inventors have done this for approximately 3,275 DIC embryos. This set of 3,275 embryos was taken from a database of information related to approximately 17,000 DIC embryos. The embryos from the database of 17,000 DIC embryos that were not included in the reduced set of 3,275 embryos used to test the model were those for which the quality of the annotations for relevant developmental events was considered unreliable, those for those in which the patient was over 40 years of age, those unrelated to day three transfers (as these particular parameters and values used for this specific example of the model are selected to classify embryos for day three transfer) and those associated with genetic screening prior to implantation. The remaining 3,275 DIC embryos comprised embryos that could be divided into four main environmental groups, namely (i) classical IVF fertilization and low oxygen incubation atmosphere, (ii) classical IVF fertilization and ambient oxygen incubation atmosphere, (iii) fertilization by IICe and atmosphere low in oxygen and (iv) fertilization by IICE and ambient oxygen atmosphere. The data for the 3,275 embryos come from 23 different clinics, with 21 clinical data for at least 10 DIC embryos and 9 of these data for at least 100 DIC embryos.

The processing of Figure 5 was applied to each of the 3,275 DIC embryos. The results of this, for example, regarding how many embryos the different scores were assigned, can be represented by a classification tree for the model, as depicted in Figure 6.

Therefore, figure 6 is a classification tree representing the application of the model of figure 5 to the 3,275 DIC embryos. Each branch node (where a ranking decision is made) and end / leaf node (where a score is assigned) is identified in Figure 6 by the corresponding stage of Figure 5 processing to which the node is related. Each node is also associated with an indication of the number of embryos (n) that reached that node of the tree and the implantation success of those embryos (imp). Therefore, the tree begins in stage S5 with a population of 3,275 embryos, of which 25.0% were implanted successfully. Stage S5 processing divides the 3,275 embryos into embryos showing direct division (267 embryos of which 5.6% were successfully implanted) and embryos not showing direct division (3,008 embryos of which 26.6% were implanted successfully). The S7 stage processing then divides the 3,008 S5 stage embryos into embryos showing slow development (470 embryos of which 12.1% were successfully implanted) and embryos not showing slow development (2,538 embryos of which 29 , 4% were implanted successfully). The S9 stage processing then divides the 2,538 S7 stage embryos into 161 embryos of which 15.5% were successfully implanted and 2229 embryos of which 30.7% were successfully implanted). It should be noted that not all embryos are classified at this stage (ie, 161 2,229 <2,538). This is because there are some embryos for which the classification criteria could not be applied, for example, because the relevant times were not identifiable in the fast-camera data. These may be termed imputed data and are deducted from further considerations in applying the model in Figure 5 to DIC embryos for validation purposes as depicted in the classification tree in Figure 6 (i.e. not assigned a score). . If there are embryos that cannot be classified in a practical application of the model to a cohort of embryos from a patient, conventional input from the embryologist may be used instead to help classify these embryos. More generally, it will be recognized that, in any event, it is to be expected that the classification according to the embodiments described herein will normally be used to assist an embryologist in making a final classification decision and the classifications obtained according to embodiments of the invention could not be taken as final without additional input from the embryologist.

As can be seen from the classification tree in figure 6, applying the model in figure 5 to DIC embryos classifies 1294 embryos with a score of 5 and these embryos have an implantation success of 36.1%. This is approximately 50% higher than the implantation success of the population as a whole, demonstrating the strength of the model in being able to identify relatively high quality / developmental potential embryos within a plurality of embryos to classify. It can also be seen that a total of 935 embryos are assigned a score of 4, with the breakdown being 410 embryos at stage S14 with an implantation success of 22.9% and 525 embryos at stage S12 with a success of implantation of 23.2%. Because embryos reaching stage S12 and embryos reaching stage S14 have been statistically found to have comparable implantation success rates, these two groups are assigned the same score.

The result of applying the model in Figure 5 to the 3,275 DIC embryos used for validation is also represented in Figure 7. This is a bar graph showing the percentage distribution of the 3,275 DIC embryos associated with the classification tree of Figure 6 between the different scores (identified by the legend "all DIC data") and also the percentage distribution of the embryos associated with the classification tree of Figure 6 that was successfully implanted (identified by the legend "DIC positive "). Also shown in figure 7 is a percentage distribution of a population of embryos known to give rise to live birth among the scores determined for these embryos according to the model represented in figures 5 and 6 (identified by the legend " live birth ").

The embryo population associated with the live birth data in Figure 7 does not simply correspond to the 3,275 DIC embryos that gave rise to a live birth. This is because live birth data are not available for these 3,275 DIC embryos. This is because once a patient becomes pregnant it does not always happen that the IVF clinic where she was treated receives information about the final outcome of the pregnancy. Of the 3,275 DIC embryos associated with the classification tree in Figure 6, it is known that only approximately 180 resulted in a live birth. It is to be expected that many other implantation embryos will also give rise to a live birth, but the data for these embryos is simply not available. Thus, the live birth data represented in Figure 7 correspond to the approximately 180 of the 3,275 embryos for which there is live birth data and another approximately 120 additional embryos for which there is live birth data.

Figure 7 demonstrates how evaluating developmental potential according to embodiments of the invention helps to identify embryos that have good developmental potential. For example, it is clear that the distributions of associated embryos with good developmental potential (ie "DIC positive" and "live birth" embryos) are skewed to higher scores than the population as a whole.

It will be appreciated that there are various modifications of the approach of Figure 5 that can be adopted in accordance with different embodiments. For example, while Figure 5 depicts a generally iterative approach in which embryos are considered and scored one by one, in other implementations embryos can be scored in parallel or at least partially in parallel. For example, instead of repeating steps S2 through S16 for each embryo, in other implementations the individual steps of the classification procedure can be performed for a plurality of embryos before proceeding to the next step of the classification procedure. For example, a stage corresponding to stage S3 may be carried out for all embryos before processing proceeds to a stage corresponding to stage S5 in which an assessment can be made of whether the embryos should be considered to have experienced direct division for all embryos that were not classified with a score of 0 at the stage corresponding to stage S3.

Also, in some implementations there may not be a stage corresponding to S3. That is, in some implementations there may not be an assessment of whether an embryo should be classified based on the appearance or not of two pronuclei. In this case, the processing can be generally similar to that represented in Figure 5, but without steps corresponding to steps S3 and S4. Similarly, there may be no steps corresponding to steps S13 and S14 in some examples.

Furthermore, and as already mentioned, different parameters and different threshold values can be used for the different stages of the method. What is significant for certain embodiments of the invention is that the model takes into account direct division, slowness in development (particularly early development), and irregularity in development (when evaluating the proportion of two characters associated with different stages of development). development of embryos) by classifying embryos according to their developmental potential. There are many different parameters that are indicative of these characteristics, such as the different examples provided above. Specific parameters and associated threshold values may vary between implementations. For example, although the above approach has been found to provide a model that is universally applicable, it may be that a specific clinic wishes to develop its own model based on these approaches and in this case may discover that other parameters and / or threshold values provide differentiation. optimized of the embryo development potential for embryos grown in that clinic. In any case, the relevant parameters and values can be established from a statistical analysis of DIC embryos associated with the relevant developmental conditions. For example, in the case of seeking to establish a universally applicable model, an analysis of how different parameters and associated threshold values performed in the classification of embryos compared to the known results for the 3,275 DIC embryos previously analyzed identified the specific parameters and values. for those parameters used in the specific illustrative implementation represented by the classification tree of Figure 6.

As already mentioned, the embodiments of the invention are aimed at classifying embryos in a way that takes into account three aspects of embryonic development, specifically, whether or not the embryos undergo direct division, whether or not the embryos present relatively development. slow and whether or not the embryos have relatively irregular development. Whether or not an embryo exhibits any of these characteristics can be established from morphokinetic / morphological data associated with the development of the embryo. In the illustrative implementation associated with the classification tree of Figure 6, the evaluation of direct division is based on the parameter (t3-tPNf), the evaluation of relatively slow development is based on the parameter t3, and the evaluation of relatively slow development Irregular is based on the parameter (t5-t3) / (t5-t2). The threshold values for each parameter in this particular illustrative implementation are adopted as discussed above and as set out in Figure 6. However, as noted above, there are a number of other parameters and corresponding thresholds that can equally be used to evaluate the three developmental characteristics identified above (direct division, slow development, irregular development) to classify embryos according to other implementations.

Therefore, Figures 8, 10 and 12 are similar to, and will be understood from, Figure 6, but show classification trees that represent different models based on different parameters and / or threshold values to evaluate whether it is considered that the embryos are associated with direct division, slow development, and / or irregular development. Figures 9, 11 and 13 are corresponding bar graphs that are similar to, and will be understood from, Figure 7 for the respective classification trees depicted in Figures 8, 10 and 12. Again, these show how the Associated embryo distributions with good developmental potential determined in accordance with embodiments of the invention are skewed to scores higher than the population as a whole.

In each of Figures 8, 10 and 12, the nodes of the classification tree are identified by the method step of Figure 5 that is most closely associated with the node. However, the models represented in Figures 8, 10 and 12 do not include all the stages corresponding to the processing of the method of Figure 5 and, therefore, there is no direct correspondence between all the nodes in the classification trees of the Figures 8, 10 and 12 and the classification tree of Figure 6.

For example, according to the classification model shown in figure 8, there is no step corresponding to step S13 in the method of figure 5. That is, the classification of embryos according to the model of figure 8 it does not imply a stage based on determining whether the embryo has reached a given stage of development in a given time. Consequently, there is one less decision node in the classification tree, and consequently the maximum score for the classification model is lower (that is, 4 instead of 5). However, it will be appreciated that the specific numerical ranking obtained according to any given model is for ranking the relative developmental potential of embryos according to that model. The numerical score is not intended to be an "absolute" rating for comparison with embryos classified with different models. That is, a score of 5 in the model represented by figure 6 is interpreted as a higher ranking than a score of 4 in the model represented by figure 6, but should not necessarily be interpreted as an indication of a higher ranking than a score of 4 in the model represented by figure 8.

Also, there is no stage corresponding to step S11 of the model represented in Figure 6 in the models represented in Figures 10 and 12. This is because the evaluation of regular development (ir) in Figure 6 is based on whether the parameter (t5-t3) / (t5-t2) falls within or outside a range defined by a lower limit in step S11 and an upper limit in step S13, but the evaluation of the development (ir) is regular in the model Figure 10 is based on an evaluation of whether (t8-t5) / (t8-2) falls within or outside the range that is limited only at one extreme.

For each model, the relative scores associated with each leaf node are based on the respective implantation success for the embryos reaching that leaf node. For example, in the model of Figure 12, DIC embryos that reach the stages marked as corresponding to stage S6 (that is, classified as undergoing direct division) and S8 (that is, classified as not undergoing direct division, but are slow developing), both have comparable implantation probabilities and are therefore assigned the same score.

As can be seen from the model represented in Figures 6, 8, 10 and 12, despite relying on different specific parameters and threshold values to evaluate embryos with respect to direct division, slow development and irregular development, however, all models can classify embryos according to implantation success. For each model, the embryos assigned the highest score are associated with implantation probabilities of approximately 32% to 36%, which represents a significant improvement over implantation probability for the population of DIC embryos taken as a whole. , which is 25%. That is, the embryo classification approach based on the three characteristics identified above can identify those with a higher probability of implantation than those with a lower probability of implantation.

Figures 14 to 17 are classification tree diagrams that are similar to, and will be understood from, Figure 6. The classification trees of Figures 14 to 18 are based on the same model as the classification tree of the Figure 6 (eg, regarding the specific parameters and thresholds used to assess direct division, slow and irregular development), but show the results of applying the model to different subpopulations of the 3,275 DIC embryos.

Figure 14 shows the result of applying the model to the subset of the 3,275 DIC embryos incubated in an atmosphere of reduced oxygen and Figure 15 shows the result of applying the model to the subset of the 3,275 DIC embryos incubated in an atmosphere of ambient oxygen. It can be seen that the results of applying the model to embryos incubated in ambient and reduced oxygen atmospheres are comparable, demonstrating that the approach is capable of classifying embryos with little sensitivity as to whether the embryos are incubated under oxygen conditions. reduced or environmental and this demonstrates the universality of approaches according to embodiments of the invention.

Figure 16 shows the result of applying the model to the subset of the 3,275 DIC embryos fertilized by IICE and Figure 17 shows the result of applying the model to the subset of the 3,275 DIC embryos fertilized by classical IVF. It can be seen that the results of applying the model to IICE embryos and IVF embryos are comparable, demonstrating that the approach is capable of classifying embryos with little sensitivity as to whether the embryos are fertilized by ICSI or IVF and this further demonstrates the universality of approaches according to embodiments of the invention.

It will be appreciated that, in addition to evaluating embryos for direct division, developmental slowness and irregularity in establishing their classification, another aspect of certain embodiments of the invention is to recognize which of these characteristics would contribute most to a low classification. In particular, direct division has been identified as a stronger indicator of low development potential than slow development, which in turn is a stronger indicator of low development than irregular development. Consequently, according to certain embodiments, embryos that exhibit direct division are ranked lower than embryos. that do not present direct division, regardless of whether the embryos meet the other criteria. That is, of the three criteria: direct division, slowness and irregularity, the determination with respect to direct division has a greater influence on the classification of an embryo in relation to the other embryos than the determination with respect to slowness and determination With regard to slowness in turn has a great influence on the classification of an embryo in relation to the other embryos than the determination with regard to irregularity.

The various classification trees discussed above for DIC embryos demonstrate how approaches for classifying a plurality of embryos in accordance with embodiments of the invention can help to identify the embryos that have the highest developmental potential (best quality). This has been shown mainly in the context of development potential measured by the probability of implantation. However, and as already noted, this is only an illustrative measure for the developmental potential of embryos and the principle described herein can also be used to classify embryos based on other measures of development potential, for For example, the probability of reaching a blastocyst stage and / or the probability that an implanted embryo will develop to a live birth and / or the probability that an implanted embryo will develop to a stage associated with a heartbeat and / or the probability that a patient to whom the embryo is transferred will become pregnant.

It will be appreciated that the staging / classification tree approach discussed above is simply an algorithmic approach to classify embryos based on this and other algorithmic approaches may in effect result in the same classification scheme. For example, instead of classifying embryos using a decision tree, as depicted in the approach in Figures 5 and 6, to identify classifications for embryos, all embryos can be evaluated with respect to all three characteristics (direct division , slowness, irregularity) and a cumulative score can be obtained based on the result of the evaluation for each characteristic. For example, embryos that do not show direct division can be assigned a score of 100 and embryos that show direct division can be assigned a score of 0 for this scoring component. Embryos that are not slow can be assigned a score of 10 and embryos that show slow can be assigned a score of 0 for this scoring component. Embryos that show no irregularities can be assigned a score of 1 and embryos that show irregularities can be assigned a score of 0 for this scoring component. Therefore, an embryo that presents direct division, slowness and irregularity will have a cumulative score of zero, while an embryo that does not present direct division, slowness or irregularity, will have a cumulative score of 111 (maximum). Based on this approach, the embryos would be sorted in an order corresponding to that of the Figure 6 approach (except with a more refined sorting of the embryos within some of the scoring categories associated with FIG. 5).

Therefore, a method to classify embryos to indicate their developmental potential has been described. The method comprises: obtaining values for a plurality of characteristics related to the morphological development of the embryos during an observation period; determine for the respective embryos whether or not the embryo has undergone a direct division event and classify embryos that have been determined to have undergone a direct division event with a classification indicating a lower developmental potential than for embryos that are not you have determined that they have experienced a direct division event; and for embryos that have not been determined to have undergone a direct division event, determine whether or not a duration of a predefined developmental stage for the embryo exceeds a predefined threshold duration and classify embryos for which the duration is determined of the predefined stage of development exceeds the predefined threshold duration with a rating indicating a lower developmental potential than for embryos for which the duration of the predefined stage of development is not determined to exceed the predefined threshold duration; and for embryos for which the duration of the predefined developmental stage is not determined to exceed the predefined threshold duration, determine whether or not the relative durations of two predefined developmental stages for the embryo are outside a predefined range and rank the embryos for which the relative durations of two predefined developmental stages for the embryo are outside a predefined range with a ranking indicating a lower developmental potential than for embryos for which the relative durations of two predefined developmental stages for the embryo embryo are not outside the predefined range.

In some aspects, some illustrative embodiments provide a method for classifying embryos to indicate their developmental potential; the method comprising: obtaining values for a plurality of characteristics related to the morphological development of the embryos during an observation period; determining for the respective embryos a measure of whether or not the embryo underwent a direct division event; determining for the respective embryos a measure of whether or not the duration of a predefined developmental stage for the embryo was greater than a predefined threshold duration; determining for the respective embryos a measure of whether or not the relative durations of two predefined developmental stages for the embryo are outside a predefined range; and classify embryos in such a way that a determination that an embryo underwent a direct division event contributes more to a classification indicating relatively low developmental potential than a determination that the duration of the pre-defined developmental stage for the embryo was longer than the predefined threshold duration and where a determination that the predefined developmental stage for the embryo was longer than the predefined threshold duration contributes more to a classification indicating relatively low developmental potential than a determination that the durations relative of two stages of predefined development for the embryo are outside the predefined range.

In some aspects, some other illustrative embodiments provide a method of establishing a score to indicate a developmental potential for an embryo; comprising: obtaining values for a plurality of characteristics related to the morphological development of the embryo during a period of observation; determining, as the first scoring component, a measure of whether or not the embryo underwent a direct division event; determining, as a second scoring component, a measure of whether or not the duration of a predefined developmental stage for the embryo was greater than a predefined threshold duration; determining, as a third scoring component, an indication of whether or not the relative durations of two predefined developmental stages for the embryo are outside a predefined range; and establishing a score to indicate a developmental potential for the embryo taking into account the first scoring component, the second scoring component, and the third scoring component such that the first scoring component has a greater influence on the score than the second or third scoring component and the second scoring component have a greater influence on the score than the third scoring component.

Therefore, and as discussed above, methods in accordance with the principles described herein can be used to establish a score to indicate a developmental potential for an embryo. The examples set forth above have primarily focused on providing an indication of the probability of successful implantation based on known implantation data for a sample of embryos. However, as already explained, the method is also applicable to establish scores related to other potential developmental characteristics for embryos, for example, the probability that an embryo will develop to the blastocyst stage.

In this regard, the model represented in figure 6 has been applied to a data set comprising 10,316 embryos from 2,413 patients / 37 clinics incubated for five days and for which it is known whether or not the embryos developed to the stage of blastocyst in 120 hours. In effect, this corresponds to the use of what might be termed "known blastocyst data" rather than "known implantation data" used for some of the other results described herein to establish the ability of the model to establish an indicator of the development potential.

The data set of 10,316 embryos was completely independent of the data set in which the model in Figure 6 was developed. This data set of 10,316 embryos contained kinetic information of the relevant developmental parameters annotated according to the guidelines proposed in the document "Proposed guidelines on the nomenclature and annotation of dynamic human embryo monitoring by a time-lapse user group" by Ciray et al. - Hum. reprod. 2014; 29; 2650-2660 [1]. The kinetic information up to day three after fertilization was used to score the embryos according to the algorithm represented by figure 6, thus assigning each embryo to one of the scoring groups (1, 2, 3, 4 or 5).

The number of embryos assigned a score of 1 was 2,024 (of which 1,911 were fertilized by traditional IVF; 93 were fertilized by ICSI; and the method of fertilization for 20 was unknown). The number of embryos assigned a score of 2 was 1,443 (of which 1,281 were fertilized by traditional IVF; 143 were fertilized by ICSI; and the method of fertilization for 19 was unknown). The number of embryos to which score 3 was assigned was 656 (of which 629 were fertilized by traditional IVF; 16 were fertilized by ICSI; and the method of fertilization for 11 was unknown). The number of embryos assigned a score of 4 was 1734 (of which 1546 were fertilized by traditional IVF; 147 were fertilized by ICSI; and the method of fertilization for 41 was unknown). The number of embryos assigned a score of 5 was 4459 (of which 3991 were fertilized by traditional IVF; 417 were fertilized by ICSI; and the method of fertilization for 51 was unknown). Therefore, for the total of 10,316 embryos, 9,358 were fertilized by traditional IVF; 816 were fertilized by IICE; and the method of fertilization for 142 was unknown.

To assess the ability of the model algorithm as a blastocyst prediction tool, the proportion of embryos that developed to the blastocyst stage (blastocyst formation) 120 hours after fertilization was determined for each of the five scoring groups. The results of this are presented in Figure 18. This clearly shows an increasing proportion of embryos associated with each score developed to the blastocyst stage with increasing score (for example, more than 60% of the embryos associated with score 5 developed to the blastocyst stage compared to approximately 10% of score 1 associated embryos that developed to the blastocyst stage). This further demonstrates the ability of the methods described above to establish an indicator of an embryonic development potential (in this case, the probability of developing to the blastocyst stage).

To assess the differences in the fraction of embryos that formed blastocysts between the scoring groups to which they were assigned, a generalized regression of linear mixed effects (RGEML, with a Bernoulli error distribution) was performed with respect to the probability of formation. of blastocysts for the different scoring groups, fertilization method (ICSI or IVF) and oxygen level incubations (reduced or ambient) as explanatory variables (fixed effects). All the interaction terms between the scoring group, fertilization method and oxygen level were included to verify if the blastocyst formation fractions in the groups Scoring were influenced by these incubation characteristics. To account for patient and clinic variability, this information was included as a random intersection, in which patients were housed in clinics. Backward elimination in stages was used to reduce the model, starting with the complete model, including all interactions of the parameters. An inclusion criterion of p <0.01 was used in the elimination procedure. The final model included the main effects of the score and the method of fertilization (IICE / IVF).

This analysis demonstrated 1) a significant difference in the probability of blastocyst formation between the scoring groups, with an increasing score being associated with a higher probability of blastocyst formation and 2) the proportion of blastocysts for IVF was higher than for ICSI. This illustrates that the principal described herein can be used based on the information available up to day three after fertilization to predict the probability of blastocyst formation. Since none of the interaction terms were found to be significant, it can be concluded that the general pattern of blastocyst proportions in the scoring groups is the same between IVF and ICSI. The fact that the proportion of blastocysts was higher in IVF than in ICSI can be explained by general clinical practice that ICSI is used especially for the most difficult cases.

Additional particular and preferred aspects of the present invention are set forth in the attached independent and dependent claims. It will be appreciated that the features of the dependent claims may be combined with features of the independent claims in combinations other than those explicitly set forth in the claims.

References

[1] Ciray et al., "Proposed guidelines on the nomenclature and annotation of dynamic human embryo monitoring by a time-lapse user group", Hum. reprod. 2014; 29; 2650-2660.

Claims (15)

1. A computer-implemented method for classifying IVF embryos to predict their developmental potential after transfer; understanding the method: obtaining values for a plurality of characteristics related to the morphological development of the embryos during an observation period; determine for the respective embryos a measure of whether or not the embryo has undergone a direct division event and classify embryos that have been determined to have undergone a direct division event with a classification indicating a lower developmental potential than for embryos that they have not been determined to have experienced a direct division event; and for embryos that have not been determined to have undergone a direct division event, determine whether or not a measure of the duration of a predefined developmental stage for the embryo exceeds a predefined threshold duration and classify embryos for which it is determined that the duration of the predefined stage of development exceeds the predefined threshold duration with a rating indicating a lower developmental potential than for embryos for which the duration of the predefined stage of development is not determined to exceed the predefined threshold duration; and for embryos for which the duration of the predefined stage of development is not determined to exceed the predefined threshold duration, determine whether a measure of a relative duration of a predefined first stage of development for the embryo relative to a second stage of development predefined for the embryo is or is not outside a predefined range and classify embryos for which the relative duration is outside the predefined range with a classification indicating a lower developmental potential than for embryos for which the relative duration is not outside of the predefined interval. The method of claim 1, wherein determining a measure of whether an embryo has undergone a direct division event comprises determining whether a first parameter associated with embryo development is less than a first threshold amount and, if so, determine that the embryo has undergone a direct division event. 3. The method according to claim 2, wherein: (i) the first parameter is a measure of the period of time between the fading of the pronuclei, tPNf, and the division time to 3 cells, t3, and where the first threshold quantity is selected from the group comprising: 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours, 11 hours, 12 hours, 13 hours and 14 hours; or (ii) the first parameter is a measure of the time period between the division time up to 2 cells, t2, and the division time up to 3 cells, t3, and the first threshold quantity is selected from the group comprising: 1 hour , 2 hours, 3 hours, 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours and 11 hours; or (iii) the first parameter is a measure of the time period between the division time up to 4 cells, t4, and the division time up to 5 cells, t5, and wherein the first threshold quantity is selected from the group comprising: 0 , 1 hours, 1 hour, 2 hours, 3 hours, 4 hours, 5 hours, 6 hours, 7 hours, 8 hours, 9 hours, 10 hours and 11 hours; or (iv) the first parameter is a measure of a ratio of a division time to 3 cells, t3, with respect to a time of division to 4 cells, t4, and wherein the first threshold quantity is selected from the group comprising: 0.9, 0.8, 0.7 and 0.6. The method according to any preceding claim, wherein determining whether a measure of the duration of a predefined developmental stage for the embryo exceeds or does not exceed a predefined threshold duration comprises: (i) determine if a measure of a division time to 2 cells, t2, exceeds a time selected from the group comprising: 27 hours, 28 hours, 29 hours, 30 hours, 31 hours, 32 hours, 33 hours, 34 hours and 35 hours; or (ii) determine whether a measure of a division time to 3 cells, t3, exceeds a time selected from the group comprising: 38 hours, 39 hours, 40 hours, 41 hours, 42 hours, 43 hours, 44 hours, 45 hours, 46 hours, 47 hours and 48 hours; or (iii) determine whether a measure of a division time to 4 cells, t4, exceeds a time selected from the group comprising: 41 hours, 42 hours, 43 hours, 44 hours, 45 hours, 46 hours, 47 hours, 48 hours , 49 hours and 50 hours; or (iv) determine if a measure of a division time to 5 cells, t5, exceeds a time selected from the group comprising: 55 hours, 56 hours, 57 hours, 58 hours, 59 hours, 60 hours, 61 hours, 62 hours and 63 hours. The method according to any preceding claim, wherein in determining whether a measure of the relative duration of a predefined first stage of development for the embryo with respect to a second predefined stage of development for the embryo is outside of a predefined range, (i) the first predefined developmental stage is a measure of the time period between the time of division to 3 cells, t3, and the time of division to 5 cells, t5, and the second predefined developmental stage is a measure of the period time between the division time up to 2 cells, t2, and the division time up to 5 cells, t5, so that the relative duration is a measure of (t5-t3) / (t5-t2), and where the predefined interval is selected from the group comprising: 0.1 to 0.9, 0.2 to 0 .8, 0.3 to 0.7, 0.4 to 0.6, and 0.5 to 0.6; or (ii) the first predefined developmental stage is a measure of the time period between the division time to 2 cells, t2, and the division time to 3 cells, t3, and the second predefined developmental stage is a measure of the period time between the division time to 2 cells, t2, and the division time to 5 cells, t5, so that the relative duration is a measure of (t3-t2) / (t5-t2), and where the Predefined range is selected from the group comprising: 0.1 to 0.9, 0.1 to 0.8, 0.2 to 0.7, 0.3 to 0.6, or 0.4 to 0.5; or (iii) the first predefined developmental stage is a measure of the period of time between the division time to 2 cells, t2, and the division time to 3 cells, t3, and the second predefined developmental stage is a measure of the period time between the division time to 3 cells, t3, and the division time to 5 cells, t5, so that the relative duration is a measure of (t3-t2) / (t5-t3), and where the Predefined range is selected from the group comprising: 0.05 to 10, 0.1 to 9, 0.15 to 8, 0.2 to 7, 0.25 to 6, 0.3 to 7, 0.35 to 6 0.4 to 5, 0.45 to 4, 0.5 to 3, 0.6 to 2, and 0.75 to 1; or (iv) the predefined first stage of development is a measure of the time period between the division time up to 3 cells, t3, and the division time up to 5 cells, t5, and the second predefined stage of development is a measure of time of division up to 5 cells, t5, so that the relative duration is a measure of (t5-t3) / t5), and where the predefined interval is selected from the group comprising: more than 0.1, more than 0, 2 and more than 0.3; or (v) the first predefined developmental stage is a measure of the time period between the division time to 3 cells, t3, and the division time to 4 cells, t4, and the second predefined developmental stage is a measure of the period time between the division time to 2 cells, t2, and the division time to 3 cells, t3, so that the relative duration is a measure of (t4-t3) / (t3-t2), and where the Predefined range is selected from the group comprising: less than 0.1, less than 0.2, less than 0.3, less than 0.4, and less than 0.5; or (vi) the first predefined developmental stage is a measure of the time period between the division time to 5 cells, t5, and the division time to 8 cells, t8, and the second predefined developmental stage is a measure of the period time between the division time up to 3 cells, t3, and the division time up to 5 cells, t5, so that the relative duration is a measure of (t8-t5) / (t5-t3), and where the Predefined range is selected from the group comprising: less than 0.1, less than 0.15, and less than 0.2; or (vii) the predefined first stage of development is a measure of the combined time periods between the division time up to 2 cells, t2, and the division time up to 3 cells, t3, and between the division time up to 4 cells, t4, and the division time to 5 cells, t5, and the predefined second developmental stage is a measure of the time period between the division time to 4 cells, t4, and the division time to 8 cells, t8, of so that the relative duration is a measure of ((t3-t2) + (t5-t4)) / (t8-t4), and where the predefined interval is selected from the group comprising: more than 0.3, more than 0.4, more than 0.5, more than 0.6, more than 0.7 and more than 0.8; or (viii) the predefined first stage of development is a measure of the time period between the division time up to 5 cells, t5, and the division time up to 8 cells, t8, and the second predefined stage of development is a measure of the period time between the division time to 4 cells, t4, and the division time to 8 cells, t8, so that the relative duration is a measure of (t8-t5) / (t8-t4), and where the Predefined range is selected from the group comprising: more than 0.3, more than 0.4, more than 0.5, more than 0.6, more than 0.7, more than 0.8, more than 0.9 and more than 0.97; or (ix) the predefined first stage of development is a measure of the time period between the fading of the pronuclei, tPNf, and the time of division to 3 cells, t3, and the second predefined stage of development is a measure of the time period between pronuclei fading, tPNf, and division time to 4 cells, t4, so that the relative duration is a measure of (t3-tPNf) / (t4-tPNf), and where the predefined interval is selected from the group comprising: more than 0.35, more than 0.45, more than 0.55, more than 0.65, more than 0.75, more than 0.85 and more than 0.95; or (x) the first predefined developmental stage is a measure of the time period between the division time to 3 cells, t3, and the division time to 4 cells, t4, and the second predefined developmental stage is a measure of the period time between the division time to 2 cells, t2, and the division time to 4 cells, t4, so that the relative duration is a measure of (t4-t3) / (t4-t2), and where the Predefined range is selected from the group comprising: less than 0.3, less than 0.4, less than 0.5, less than 0.6, and less than 0.7; or (xi) the first predefined developmental stage is a measure of the time period between the division time to 5 cells, t5, and the division time to 8 cells, t8, and the second predefined developmental stage is a measure of the period time between the division time up to 2 cells, t2, and the division time up to 8 cells, t8, so that the relative duration is a measure of (t8-t5) / (t8-t2), and where the Predefined range is selected from the group comprising: less than 0.2, less than 0.3, less than 0.4, less than 0.5, and less than 0.6. 6. The method according to any preceding claim, further comprising: for embryos for which the relative duration of the predefined first stage of development with respect to the predefined second stage of development is outside a predefined range, determining whether the relative duration is above or below the predefined range; and classify embryos for which the relative duration is outside the predefined interval on one side of the predefined interval with a classification indicating a lower developmental potential than for embryos for which the relative duration is outside the predefined interval on the other side of the predefined interval. The method according to any preceding claim, further comprising for embryos that have been determined to have undergone a direct division event, determine whether or not a measure of the duration of a predefined developmental stage for the embryo exceeds a predefined threshold duration and classify embryos that have been determined to have undergone a direct division event and for which the duration of the predefined stage of development is determined to exceed the predefined threshold duration with a rating indicating lower developmental potential than for embryos that have been determined to have undergone a division event direct and for which the duration of the predefined stage of development is not determined to exceed the predefined threshold duration. The method according to any preceding claim, further comprising for embryos that have been determined to have undergone a direct division event, determining whether or not a measure of a relative duration of a predefined first developmental stage for the embryo relative to a second predefined developmental stage for the embryo is out of a predefined interval and classify embryos that have been determined to have undergone direct division and for which the relative duration is outside the predefined interval with a classification indicating a lower developmental potential than for embryos that have been determined to have experienced a direct split event and for which the relative duration is not outside the predefined range. The method according to any preceding claim, further comprising For embryos that have not been determined to have undergone a direct division event and for which the duration of the predefined stage of development is determined to exceed the predefined threshold duration, determining whether a measure of a relative duration of a first stage of predefined development for the embryo with respect to a second stage of development predefined for the embryo is outside a predefined interval and classify embryos that have been determined to have undergone direct division and for which the duration of the stage of The predefined developmental duration exceeds the predefined threshold duration and for which the relative duration is outside the predefined range with a rating indicating a lower developmental potential than for embryos that have been determined to have undergone a direct division event and for which it is determines that the duration of the predefined developmental stage exceeds the to predefined threshold duration and for which the relative duration is not outside the predefined range. 10. The method according to any preceding claim, wherein, For embryos for which the relative duration of the predefined first stage of development with respect to the predefined second stage of development has been determined to be within the predefined range, determining whether or not the respective embryos have developed to a predefined number of cells within a predefined time and classify embryos that have not developed to the predefined number of cells in the predefined time with a classification indicating a lower development potential than for embryos that have been determined to have developed to the number predefined cells at the predefined time. The method according to any claim 10, wherein the predefined number of cells is eight cells and the predefined time is selected from the group comprising: 64 hours, 65 hours, 66 hours, 67 hours, 68 hours, 69 hours, 70 hours , 71 hours and 72 hours. The method according to any preceding claim, further comprising issuing an indication representing the classifications for at least some of the embryos against each other. 13. An apparatus (2) for classifying IVF embryos to predict their developmental potential after transfer, the apparatus comprising: a data input element (6) configured to obtain values for a plurality of characteristics related to the morphological development of the embryos during an observation period; and a processor element (4) configured to: determine for the respective embryos a measure of whether or not the embryo has undergone a direct division event and classify embryos that have been determined to have undergone a direct division event with a classification indicating a lower developmental potential than for embryos that they have not been determined to have experienced a direct division event; and for embryos that have not been determined to have experienced a direct division event, determine whether or not a measure of the duration of a predefined developmental stage for the embryo exceeds a predefined threshold duration and classify embryos for which the duration of the predefined stage of development is determined to exceed the predefined threshold duration with a classification indicating a lower developmental potential than for embryos for which the duration is not determined of the predefined stage of development exceeds the predefined threshold duration; and for embryos for which the duration of the predefined stage of development is not determined to exceed the predefined threshold duration, determine whether a measure of a relative duration of a predefined first stage of development for the embryo relative to a second stage of development predefined for the embryo is or is not outside a predefined range and classify embryos for which the relative duration is outside the predefined range with a classification indicating a lower developmental potential than for embryos for which the relative duration is not outside of the predefined interval. 14. A non-transient computer program product containing machine-readable instructions which, when implemented in a computer, causes the computer to carry out the method of any of claims 1 to 12. 15. An apparatus (4) loaded with and operative for executing machine-readable instructions for carrying out the method of any of claims 1 to 12.